Blog @ Illumina
Real scientists. Real commentary.

Technology and Applications at AGBT 2013

Abizar Lakdawalla, Ph.D.
| Feb 28, 2013

Gears 3131230

What about new technology at AGBT? Quite uncharacteristically, there were fewer grand announcements this year. Some of the technologies presented in previous years came back in different incarnations. For applications, Jay Shendure’s update on Molecular Inversion probes seem to nicely fit the need for sequencing of fewer genomic regions in very large sample cohorts at low cost. Sunney Xie gave more details on the MALBAC protocol for the amplification of genomic DNA from single cells, along with some exciting application examples. On the sequencing platform side, Starlight single-molecule sequencing was reincarnated as a detector of methylation in single cells. Intelligent Bio-Systems’ sequencing platform re-appeared as the GeneReader. The only platform that I had not heard about before was PicoSeq. More details below.

Jay Shendure from the University of Washington described targeted sequencing with a new version of molecular inversion probes (MIPs). MIPs use probes with two locus-specific regions at the ends that form a partial circle when hybridized to the target region. The gap in the circle is filled by extension and ligation. Exonucleases remove non-circularized DNA to achieve a high signal-to-noise ratio. The new MIP 2.0 approach incorporates barcodes and can be used for up to 10,000 targets, providing a cost-effective strategy for large cohort studies with intermediate levels of gene targets at < $1 per gene per sample. By applying MIP 2.0 for sequencing of 44 candidate genes in 2,364 trios, 27 de novo protein altering events were found, six of which were not detected by exome sequencing. MIPs are currently being applied to autism, intellectual disability, cancer, and a 10,000-sample study on hemophilia.

Sunny Xie from Harvard University gave an interesting talk on single-cell genomics based on a single-cell genomic DNA amplification protocol, MALBAC (multiple annealing and looping-based amplification cycles; Science 338, 1622. 2012) that gave 93% genome coverage at 30× sequencing depth. He showed CNV data from multiple circulating tumor cells from 5 mL of blood from five lung cancer patients. Reproducible CNVs and point mutations were detected. In another example, the cross-over points for chromosomes were determined by sequencing of individual sperm cells. His group used MALBAC for preimplantation genomic screening and achieved higher accuracy and coverage by sequencing the first and second polar bodies at low depth for chromosomal abnormality detection.

Chongyuan Luo from Joe Ecker’s lab at the Salk Institute evaluated Life Technologies’ single-molecule approach, Starlight, for the analysis of DNA methylation in single cells. Starlight uses FRET between a labeled-DNA polymerase and fluorescently-labeled dNTP. A total internal reflection microscope is used to sequence synthetic templates and managed to get up to 96% accuracy for 26 bp, although the polymerase seemed to fail in the first five bases for most molecules. The median accuracy was 80%. Rarely, read length reached 61 bp (I think it was for 5–10% of all reads). The lab is working on improving image processing and, as Starlight needs real-time imaging, they are building a device with photoactuated droplet microfluidics (PDMs) to allow sequential activation of one tile at a time to increase data output.

Intelligent Bio-Systems’ PinPoint Mini re-appeared as the GeneReader from Qiagen. Dirk Leoffert from Qiagen focused on the workflow requiring two modified QIAcube automation systems to serially prepare libraries and perform bead-based clonal amplification. The beads are then loaded into single-lane flow cells. The GeneReader holds 20 of these flow cells with the potential to produce 2 GB per flow cell. Read lengths and quality metrics were not presented. Qiagen is collaborating with SAP for data analysis.

Gordon Hamilton talked about PicoSeq technology, which measures the length of single DNA molecules from interference rings. Single DNA molecules are attached to a solid substrate and the other end tagged with a magnetic bead. Under a magnetic field, the DNA is stretched upwards and the distance to the magnetic bead is estimated from optical interference rings. The method called SIMDEQ (SIngle-molecule Mechanical DEtection and Quantification) uses changes in the hairpin loop of a 2.5-Kb fragment induced by the binding of querying oligos. Apparently, multiple probes can be used simultaneously for diverse applications including DNA fingerprinting, triplet repeat diseases, or bacterial phylogenetics. For sequencing, a modified version for sequencing ligation would be used. Currently 50 fragments can be queried in parallel, and the goal is to move to one million.

Finally, there was some buzz about Illumina’s new synthetic long-read Moleculo Technology with two excellent posters on genome phasing from co-founder Mickey Kertesz and on de novo haplotype resolution from Dmitry Pushkarev. Applications for Moleculo include de novo assembly of non-model organisms, metagenomics sequencing projects, detailed analysis of unstable cancer genomes, and distinction between functional genes and closely related pseudogenes.